**Proof plan.** The proof proceeds by contradiction and exhaustion of alternatives. The $\omega$-[limit](/page/Limit) set $\omega(\gamma)$ of the trajectory is shown to be non-empty and compact (by the trapping hypothesis), and the Poincaré-Bendixson theory then classifies its structure: in $\mathbb{R}^2$, a non-empty compact $\omega$-limit set that contains no equilibrium must be a periodic orbit. The key ingredient is the Jordan Curve Theorem, which is specific to two dimensions and is why the result fails in $\mathbb{R}^3$. **Step 1 (Non-emptiness and compactness of $\omega(\gamma)$).** The $\omega$-limit set is defined as \begin{align*} \omega(\gamma) := \bigcap_{T \geq 0} \overline{\{\gamma(t) : t \geq T\}}. \end{align*} Since $\gamma(t) \in R$ for all $t \geq 0$ and $R$ is compact, $\omega(\gamma)$ is a nested intersection of non-empty compact [sets](/page/Set), hence non-empty and compact. Moreover, $\omega(\gamma) \subseteq R$. **Step 2 (Invariance).** The $\omega$-limit set is positively invariant under the flow: if $p \in \omega(\gamma)$, then there exists a [sequence](/page/Sequence) $t_k \to \infty$ with $\gamma(t_k) \to p$. By continuous dependence on initial data, $\gamma(t_k + s) \to \varphi_s(p)$ for any fixed $s > 0$, and since $t_k + s \to \infty$, we have $\varphi_s(p) \in \omega(\gamma)$. The same argument with backward time (using the fact that $\omega(\gamma)$ is compact and $R$ has no equilibria to block backward extension) shows that $\omega(\gamma)$ is in fact invariant (both forward and backward). **Step 3 (Structure of $\omega(\gamma)$ without equilibria).** Since $R$ contains no equilibria by hypothesis, neither does $\omega(\gamma) \subseteq R$. For any point $p \in \omega(\gamma)$, the orbit $\{\varphi_t(p) : t \in \mathbb{R}\}$ through $p$ lies entirely in $\omega(\gamma)$. [claim:Periodic Orbit In Omega Limit Set] If $\omega(\gamma)$ contains no equilibrium, then $\omega(\gamma)$ is a single periodic orbit. [/claim] [proof] Take $p \in \omega(\gamma)$ and construct a local cross-section $\Sigma$ — a small line segment through $p$ transversal to the flow (possible since $f(p) \neq 0$). The trajectory $\gamma$ must cross $\Sigma$ infinitely often (since $p \in \omega(\gamma)$). Let $p_1, p_2, \ldots$ be successive crossings. The Poincaré map $P: \Sigma \to \Sigma$ defined by following the flow from one crossing to the next is a well-defined continuous map. The sequence $(p_k)$ has a subsequence converging to some $q \in \Sigma \cap \omega(\gamma)$. The orbit through $q$ is contained in $\omega(\gamma)$ and returns to $\Sigma$. If $P(q) \neq q$, then the orbit segment from $q$ to $P(q)$, together with the segment of $\Sigma$ between them, bounds a region. By the Jordan Curve Theorem, this region is simply connected, and its [boundary](/page/Boundary) is a simple closed curve. But then the flow must re-enter this region through $\Sigma$ (since the flow is transversal there), contradicting uniqueness of the Poincaré map. Therefore $P(q) = q$, meaning the orbit through $q$ is periodic. Finally, $\omega(\gamma)$ cannot contain any point not on this periodic orbit: if it did, the orbit through that point would also be contained in $\omega(\gamma)$, and by the Jordan Curve Theorem applied to the periodic orbit (which is a simple closed curve in $\mathbb{R}^2$), such an orbit would have to lie entirely inside or outside the periodic orbit. In either case, [continuity](/page/Continuity) of the flow and the definition of $\omega$-limit set lead to a contradiction with the non-crossing property of trajectories in $\mathbb{R}^2$. [/proof] This completes the proof: $\omega(\gamma)$ is a single periodic orbit contained in $R$.